Modeling Bimodal Discrete Data Using Conway-Maxwell-Poisson Mixture Models

Bimodal truncated count distributions are frequently observed in aggregate survey data and in user ratings when respondents are mixed in their opinion. They also arise in censored count data, where the highest category might create an additional mode. Modeling bimodal behavior in discrete data is useful for various purposes, from comparing shapes of different samples (or survey questions) to predicting future ratings by new raters. The Poisson distribution is the most common distribution for fitting count data and can be modified to achieve mixtures of truncated Poisson distributions. However, it is suitable only for modeling equidispersed distributions and is limited in its ability to capture bimodality. The Conway–Maxwell–Poisson (CMP) distribution is a two-parameter generalization of the Poisson distribution that allows for over- and underdispersion. In this work, we propose a mixture of CMPs for capturing a wide range of truncated discrete data, which can exhibit unimodal and bimodal behavior. We present methods for estimating the parameters of a mixture of two CMP distributions using an EM approach. Our approach introduces a special two-step optimization within the M step to estimate multiple parameters. We examine computational and theoretical issues. The methods are illustrated for modeling ordered rating data as well as truncated count data, using simulated and real examples.     

What Affects the Quality of Economic Analysis for Life‐Saving Investments?‡

Economic analysis of life-saving investments in both the public and private sectors has the potential to dramatically improve longevity and the quality of life, but only if the analyses on which decisions are based are done well. In this article, we analyze a data set that provides information on the content and quality of journal articles that measure the cost-effectiveness of life-saving investments. Our study is the first to provide a detailed multivariate analysis of factors affecting objective measures of quality. We also explore whether a series of recommendations by an expert panel convened by the U.S. Public Health Service affect the way analyses of specific life-saving investments are done. Our results suggest that four factors are positively correlated with an index we construct to measure analytical quality: (1) having at least one author affiliated with a university, (2) publication in a journal that has experience in publishing these analyses, (3) if the life-saving investment is located in the United States, and (4) if the analysis considers a measure of social costs or benefits. Somewhat surprisingly, a study's funding source and whether it is affiliated with industry are not significantly correlated with the quality index. Finally, neither time nor the panel guidelines had an impact on the index.     

Fairness and voting

In this paper we present a two-period model in which we examine how concern about fairness might affect voter behavior. We show that in the first period politicians choose the median voter’s position even if this does not correspond to their bliss points and neither they nor the voters can commit to a particular action. Moreover, concern about fairness creates substantial incumbency advantages. Our results hold even if voters care very little about fairness.     

Racial segregation and the limits of international undergraduate student diversity

This study challenges the assertion that the influx of Asian international undergraduate students in universities across the United States creates richer educational and social environments. Drawing on qualitative research at a public university with a large number of Asian international students, this article examines how Asian American student leaders and their organization took on the difficult institutional task of actualizing the diversity of these new students in a racially segregated campus. We found that instead of viewing racial segregation practices as possibly tied to racial discrimination and privileges of normative whiteness, students expressed both support and resistance to Asian international students in race-neutral language of comfort and organizational differences that reflects the dominant ideology of colorblindness. We argue that any claims to the benefits of international student diversity must take serious account of colorblind racism and the experiences of racial marginalization and racial segregation among domestic minority and international students.     

Tax rules

Tax schemes are more restricted by constitutional rules than subsidies. We introduce a model of public good provision with incentive problems in agenda-setting and identify several advantages of restrictions on tax schemes. In particular, tax rules may prevent the proposal and adoption of inefficient projects that benefit only a small minority. We also propose “redistribution efficiency” as a socially desirable property of proposals and find that tax rules guarantee this kind of efficiency. Moreover, a constitution that requires certain restrictions on both taxes and subsidies satisfies a criterion we label “robustness against counter-proposals.” We provide an axiomatic characterization of this constitution. Finally, we perform utilitarian welfare comparisons.     

Social influences on use of cigarettes,e-cigarettes,and hookah by college students

ABSTRACT

Objectives: (1) Compare social norms and perceived peer use between college student cigarette, e-cigarette, and/or hookah users and nonusers; and (2) determine variables associated with social influences. Participants: Undergraduate students attending a large university in the Southeast United States (N = 511). Methods: An April 2013 online survey assessed use of 3 types of tobacco, social norms, perception of peer use, number of smokers in life, exposure to secondhand smoke, and demographic characteristics. Results: Participants indicated greater acceptance of emerging tobacco products than for cigarettes and consistently overestimated the percent of peers who use various tobacco products. Males and current users had higher social norm scores for all 3 forms of tobacco. Conclusion: To counter marketing of alternative tobacco products, education about the dangers of their use needs to be implemented across college campuses as part of a comprehensive tobacco control strategy that also includes tobacco-free campus policies.     

Control and Alcohol-Problem Recognition Among College Students

Abstract

Objective: This study examined negative control (ie, perceived lack of control over life outcomes) and need for control as predictors of alcohol-problem recognition, evaluations (good/bad), and expectancies (likely/unlikely) among college students. The study also explored the interaction between the need for control and alcohol consumption in alcohol-related outcomes. Participants: Participants were a convenience sample of 500 college students from a rural Midwest university. Data were collected during the 2009–2010 academic year. Methods: Participants completed a survey assessing control and alcohol-problem recognit-ion, evaluations, and expectancies. Results: Negative control demonstrated a significant positive association with alcohol-problem recognition, evaluations, and expectancies after controlling for gender and alcohol consumption. Need for control did not have a main effect. However, the interaction was significant in that the association between need for control and negative evaluation of alcohol problems was strongest among participants with the highest levels of alcohol consumption. Conclusions: These results demonstrate that individuals' differences in sense of control are associated with alcohol-problem recognition, evaluations, and expectancies in young adults.     

Counseling outside the lines: Creative arts intervention for children and adolescents,individual, small group and classroom applications by Vernon,A., and Barry,K. L.

No abstract available for this article.     

Measuring Social Well-Being in People with Chronic Illness

Although social well-being (SWB) is recognized as an integral component of health, it is rarely included in health-related quality of life (HRQL) instruments. Two SWB dimensions were identified by literature review: social support (SWB-SS) and social function (SWB-SF). As part of a larger project to develop item response theory-derived item banks and computerized adaptive testing, we developed and tested items for the SWB banks. Item ratings of three large (n > 600) datasets were conducted by 15 reviewers. Rasch measurement analyses were conducted to initially define item hierarchies. Out of 83 total items, 8 were removed due to model misfit and 8 were removed because of overlapping item content. We then wrote 11 new SWB-SS and 16 new SWB-SF items to fill content gaps, and edited items to improve comprehension and consistency. A total of 94 items (65 SWB-SS, 29 SWB-SF) was administered by computer to 202 cancer patients. Confirmatory factor analyses, Rasch analyses, and evaluations of construct validity were performed. Patients commented favorably on the content of the items and expressed appreciation for attention to this aspect of their HRQL. Using current psychometric standards for unidimensionality, reliability, and content and construct validity, we derived six preliminary item banks for social support (instrumental support, informational support, positive and negative emotional support, positive and negative social companionship) and two for social function (limitations and satisfaction). The empirical construct hierarchy was consistent with clinical observations; e.g., hobbies and leisure activities tended to reflect more limitations, while meeting the needs of family and friends tended to reflect fewer limitations. Optimal care for patients with cancer or other chronic illnesses includes obtaining a complete picture of patients’ physical and psychosocial health status. SWB measures are important since diseases like cancer and their treatment can affect quality of relationships, parental responsibilities, work abilities and social activities. With properly calibrated item banks, it will be possible to precisely and efficiently measure and monitor multiple HRQL dimensions in individual patients, and use their responses to inform care. Qualitative patient feedback and quantitative analyses suggest that it is possible and desirable to include SWB measures in HRQL assessment.